I disagree with your example, but either way your conclusion is agreeing with my point. If you can’t give a sensible answer, then don’t give a random one confid

I disagree with your example, but either way your conclusion is agreeing with my point. If you can’t give a sensible answer, then don’t give a random one confidently (like an LLM). The user in question is doing the equivalent of always answering “infinity” (the best case) to your example.